Language Matters: The Science Behind Inclusive Job Ads and Why it's Crucial for Business Growth
Diversely Co-founder and CPO, Hayley Bakker, dives into the science behind inclusive job ads and what helps make them appealing to diverse talent groups. And crucially why this is vital for business growth and health…
Globally advertising expenditure has sky-rocketed and Asia Pacific is no exception at a total spend of almost USD 200 billion last year. Companies are heavily investing both time and money into SEO and ad words, 95% of Google’s revenue comes from this source.
Why are companies spending this kind of money on advertising?Because they believe their content if formulated correctly, can and will persuade target customers to buy.
And they are right! Not only demonstrated through increasing revenues, but also backed by psychological principles stating that buyers behavior is influenced by their empathetic response - whether a person empathizes with, or feels close to, a brand. Let’s take a look at Nike’s ‘Just do it’ tagline and campaign. One of the 1988 campaign’s objectives was to target all Americans – regardless of age, gender, or physical fitness level. The success of ‘Just do it’ is often attributed to its tagline being ‘universal and intensely personal’ as opposed to many of its competitors at the time.
You may be wondering how this relates to our blog topic of today. Let’s recap:
People convert to a brand when they can empathise and feel close to it and companies put in significant time and money to get their messaging just right for their audience.
The current (tech) talent crunch in many major hubs, as well as Korn Ferry’s workforce projection, show a significant problem on the horizon: 85 million jobs globally will go unfilled by 2030. Isn’t it therefore time, right now, for companies to start rethinking hiring strategies and (rewriting) job descriptions to attract and appeal to a broader talent pool?
Hiring Competitive Edge
UNTAPPED TALENT.
Well, companies are beginning to revisit their hiring strategies and considering new ways to broaden their talent pool, either through increased exposure within their current target groups (incl. referrals) or by tapping into entirely new talent groups. Some of the greatest untapped talent pools include women, returnships, experience (40+), people with disabilities, and immigrants. So why are these groups remaining untapped in certain industries?
WORDS MATTER.
Research shows that the words people use does reveal a lot about their social and psychological state. It even reveals differences in the way ethnic groups or men and women use everyday language. Women, for example, use a more communal style of speech than men and make more references to social and emotional words. Ample evidence suggests that belongingness — feeling that one fits in with others within a particular domain — affects people’s engagement within a domain.
In summary, science tells us that:
1. Speech differs among groups of people, whether gender, minority groups, age, etc.
2. Messaging impacts a sense of empathy and belonging
3. People will engage within a domain based on a sense that one fits in
APPEALING JOB DESCRIPTION.
A review of random samples job descriptions shows that male and female themed words are differentially present in job ads that are typically associated with and attracting men versus women. Confirming the concept that language plays a role in reinforcing and perpetuating diversity inequalities. Diversely helps to hire managers and recruiters to not only identify [gendered, racial, age] themed words in their job descriptions but more importantly suggests an alternative, more inclusive language. Why?
Well-balanced and carefully worded job descriptions that appeal to a broader talent pool – or in some cases a different, more specific, talent group – will attract more (suitable) applicants and increase companies’ competitive edge when it comes to hiring.
“Work hard, play hard” instead “Work smart, play more”
“Strong track record in” instead “Demonstrable ability to”
“applicants from top tier universities” instead focus on the type of experience
“Inviting English native speakers” instead “Inviting fluent English speakers”
“He/ she will work directly with…” instead “You will work directly with”
OUR TRAINING MODELS.
At Diversely, we train and ‘feed’ our machine learning models with labeled and tagged job description data from our research partner Nottingham University’s Centre for Research in Applied Linguistics. We use sampled ‘real-life’ job descriptions from all walks of life, including varying companies, industries, and countries. University researchers have used existing scientific findings on the relationships (correlation) between phrases and demographics to analyze and tag bias (or coded) language. As our clients continue to feed new job descriptions into our analyzer, our models will continually be updated with new phrases and language.
APPEAL AND STRUCTURE SCORE.
Every job description is scored based on its appeal and structure. More on how to improve the latter with some hard-won tips in our previous blog. Diversely’s appeal score ranges from 0 to 100, where a score of 100 reveals no bias and therefore a highly inclusive job description and a score of 0 reveals highly bias language across the entire text. Below the scoring, our analyzer will show you the diversity areas in which your job description text is demonstrating bias, across gender, age, and social background. It may reveal that a bias score of 75 driven by a job description language that is biased towards young women from a high social class, for example.
How Does It Work?
BIAS CODED PHRASES.
Our ‘Diversely Job Description Analyser’ identifies and high-lights bias, also known as coded phrases, in job description texts. Before we proceed, it is important to reflect on the term ‘bias’. Bias in layman terms it is the inclination or tendency to lean in a certain direction towards a certain thing. While it has a negative connotation, bias is not by definition negative and should by no means be confused with its evil cousin’s prejudice and discrimination. Bias in this context simply suggests that a job description might lean towards, or has the tendency to appeal more to, a certain demographic.
HOW TO USE APPEAL AND BIAS TO ACHIEVE YOUR GOALS?
“Seeking to attract broader talent to your organisation? Go for an overall high appeal score
Are women under-represented in your organisation? It’s fine to have a lower appeal score, as long as your bias phrases are feminine
Is your organisation and talent pool usually very young? Again, it’s fine to have a lower appeal score, as long your bias phrases are mature/ experienced
What does the initial data show?
The numbers have been very promising so far, with a higher inclusivity score, relating to a greater volume of job applicants, a significant increase in the (gender) diversity of job applicants, and perceived higher quality of the candidate pipeline. How do we know?
At Diversley, we conducted an experiment across 20 job adverts. First we removed other factors which might impact the performance of the job advert. We did this by using the same recruiter account across all jobs, only posting the role to one and the same job board (LinkedIn), and without changing any of the actual candidate requirements in the job advert. The findings have been fascinating. The rewritten job advert - compared to the original job advert - achieved:
An average of 28% more job applicants
Up to 68% more female job applicants
Up to 2x as many (100% increase of) relevant applicants shortlisted
Another surprising outcome, came from LinkedIn’s own algorithm, recommending on average 30% more female candidate profiles for the re-written job advert, compared to essentially the same role advertised using different - less inclusive - language.
Your Next Hire
To win in the highly competitive talent market of the (near) future, companies and hiring managers will need to understand their current workforce and gaps, identify broader and untapped talent pools and update their approach and language to appeal to and attract more diverse talent. These are great first steps, but to truly become competitive, employers need to go a lot further, considering how their current teams and company culture welcome and create a sense of belonging for new employees from varying backgrounds.
